rendering
PlantUML MCP, CLI, and Lua Integration for AI Agents
PlantUML integration docs for AI agents: MCP gateway setup, PlantUML CLI commands, Lua API reference, credentials, and function schemas.PlantUML for agents
No credentials required; available in web and CLI hosts when runtime dependencies are installed.
Use this integration from Lua code mode, the headless integrations CLI, or the KosmoKrator MCP gateway. The same package metadata powers all three surfaces.
Agent Surfaces
Machine-Readable Metadata
Runtime Requirements
| Name | Type | Description |
|---|---|---|
java | binary | Java is required to run PlantUML. |
Function Catalog
| Function | Type | Parameters | Description |
|---|---|---|---|
plantuml.render_plantuml | Write write | 2 | Render a PlantUML diagram to a PNG image. Pass valid PlantUML syntax and get back a markdown image embed. Supported diagram types: class, sequence, activity, component, state, use case, object, deployment, timing, network (nwdiag), wireframe (salt), Gantt, mindmap, WBS, JSON, YAML, ERD. Example syntax: ``` @startuml Alice -> Bob: Authentication Request Bob --> Alice: Authentication Response Alice -> Bob: Another request Bob --> Alice: Another response @enduml ``` Tips: - Always wrap syntax in @startuml / @enduml - Use `->` for solid arrows, `-->` for dashed arrows - Use `class ClassName { }` blocks for class diagrams - Use `(*) -->` for activity diagram start - Use `[Component]` for component diagrams - Use `state "Name" as s1` for state diagrams |